Correspondence to: Prof. F. M. McAuliffe, UCD Obstetrics & Gynaecology, National Maternity Hospital, Dublin 2, Ireland (e-mail: email@example.com)
Selective fetoscopic laser photocoagulation (SFLP) is now the treatment of choice for twin–twin transfusion syndrome (TTTS). The incidence of recurrent TTTS following SFLP has been inconsistently reported across different studies. We performed a systematic review of TTTS recurrence following SFLP.
PubMed and MEDLINE online databases were searched for articles published between 2000 and August 2011, using combinations of the terms ‘twin–twin transfusion’, ‘TTTS’, ‘laser’, ‘recur’ and ‘outcome’. Citations identified in the primary search were screened for eligibility. Studies reporting outcomes from selective SFLP for TTTS in twin pregnancies, which addressed specifically the issue of TTTS recurrence, were included. The primary outcome was rate of TTTS recurrence. Secondary outcomes were therapeutic preference and fetal outcomes in cases of recurrent TTTS.
The primary search identified 22 eligible studies that are included in this review (n = 2447 twin pregnancies). Two studies included a minority of non-selective procedures. The published incidence of recurrent TTTS ranged from 0 to 16%. Clinical management was reported in 65.7% (71/108) cases, with repeat SFLP the most commonly performed secondary intervention. Only three studies provided comprehensive outcome data for cases of recurrent TTTS. The overall rate of neurologically-intact survival was 44% (23/52). The data were inadequate to determine the effects of secondary therapeutic approach, placental location or gestational age on perinatal outcome in cases of recurrent TTTS.
The published rate of TTTS recurrence following SFLP in monochorionic twin pregnancies ranges from 0 to 16%. Although limited follow-up data suggest that recurrence is associated with significant perinatal mortality and morbidity, further study is needed. Currently, there are insufficient data available to guide recommendations for clinical management of TTTS recurrence. Future studies on SFLP for TTTS must include details on recurrence rates and provide outcome data specific to the recurrent subset.
Twin–twin transfusion syndrome (TTTS) complicates 10–15% of monochorionic diamniotic (MCDA) twin pregnancies[1, 2] and is associated with significant perinatal morbidity and mortality. Laser photocoagulation of placental anastomotic vessels as a treatment option for TTTS was first proposed by De Lia and colleagues more than 20 years ago. Early studies reported outcomes following non-selective laser coagulation of all vessels crossing the dividing membrane[4, 5]. Operative techniques soon evolved and the concept of selective fetoscopic laser photocoagulation (SFLP), in which only true placental anastomoses at the vascular equator are obliterated, was introduced by Quintero et al. in 1998 and shown to confer benefit in 2000.
MCDA twin pregnancies are at risk of all the obstetric complications common to multiple gestations, particularly preterm delivery. In addition, MCDA twins are at risk of complications unique to monochorionic twins, predominantly TTTS. Cases treated with SFLP are at further risk of iatrogenic preterm rupture of membranes, recurrent TTTS, iatrogenic twin anemia–polycythemia sequence (TAPS) without amniotic fluid discordance and intrauterine death of one or both twins. Although the rate of laser-related twin demise has received considerable attention in the literature to date[12-14], the issue of TTTS recurrence following SFLP remains poorly documented.
A previous review by Yamamoto and Ville documented cases of TTTS recurrence reported between 1995 and 2007 and found a wide range of recurrence rates across studies. However, this review included both selective and non-selective FLP and these fetoscopic techniques are associated with differing rates of recurrence[15, 16]. Furthermore, since 2007, several authors have reported outcomes following SFLP for TTTS[12, 16-28]. Here we present a systematic review of the incidence and management of recurrent TTTS after SFLP.
This systematic review was prepared in accordance with the recommendations of the PRISMA statement. PubMed, MEDLINE and Cochrane Databases were searched in August 2011 for combinations of the terms ‘twin–twin transfusion’, ‘TTTS’, ‘laser’, ‘recur’ and ‘outcome’. The primary search was performed by the first author (C.W.) for relevant citations published between 2000 (the year that SFLP for TTTS was first introduced) and August 2011, without language restrictions. Online trial registries (controlled-trials.com and clinicaltrials.gov) were also searched. Reference lists from articles retrieved in the primary search were screened for additional citations. The corresponding authors of included studies with missing data were contacted and asked to provide additional clinical details where necessary.
Studies were eligible for inclusion if they met the following inclusion criteria: prospective or retrospective studies reporting outcomes from selective (not non-selective) fetoscopic laser photocoagulation for TTTS; TTTS affecting monochorionic twin (not triplet) pregnancies; SFLP performed at < 26 weeks' gestation; and rate of TTTS recurrence explicitly stated in study. Studies were excluded from the analysis if any one of the inclusion criteria was not met. Case reports and small case series (< 10 cases) were excluded from this review.
Diagnosis of recurrent TTTS
Recurrence of TTTS following FLP is usually defined as the finding of polyhydramnios (deepest vertical pool > 8 cm at ≤ 20 weeks' gestation or > 10 cm at > 20 weeks) or oligohydramnios (deepest vertical pool ≤ 2 cm) sequence following laser therapy[11, 30]. Some authors stipulate that recurrent TTTS be defined only in cases of recurrent disease occurring > 1 week beyond the SFLP procedure (Table 1)[11, 22, 31].
Table 1. Rates of recurrence of twin–twin transfusion syndrome: review of the literature
Followed up (n)
Recurrent twin–twin transfusion syndrome
Twin pairs (n (%))
Only first author of each study is specified.
Only cases with double twin survivors 1 week following SFLP included.
Five miscarriages and six TOPS within first week of SFLP excluded.
All cases of SFLP performed after a prior amnioreduction.
Varying levels of selectivity, although non-selective cases (prior to 2000) excluded.
22% of SFLP cases had a failed trial of prior amnioreduction (n = 33).
Cases of triplets excluded from present analysis (n = 3).
One third of cases described as being non-selective.
Cases of SFLP after 26 weeks excluded (n = 3).
Pregnancies complicated by fetal death (n = 23) excluded from analysis.
Data obtained from authors of primary studies. FLP, fetoscopic laser photocoagulation; N/A, not available; Non-sel, non-selective FLP; Pro, prospective; RCT, randomized controlled trial; Retro, retrospective; SFLP, selective FLP; TOPS, twin oligohydramnios–polyhydramnios sequence.
The primary outcome of the present review was incidence of TTTS recurrence following SFLP for TTTS in monochorionic twin pregnancies. Secondary outcomes were therapeutic preference in cases of recurrent TTTS and fetal outcomes associated with recurrent TTTS. We also assessed the relationship between placental location and rate of TTTS recurrence.
The primary search returned 228 studies with potential for inclusion. These abstracts were screened for eligible studies, yielding 61 citations (Figure 1). These 61 papers were examined in detail and their references screened for further eligible citations. Ten citations were review articles only without primary data. A further six citations did not specify a selective approach to laser photocoagulation (n = 2), were placental-based studies without clinical outcomes (n = 3) or are currently recruiting (n = 1). Of the 45 remaining studies meeting the inclusion criteria, 23 did not provide clinical data on TTTS recurrence and were excluded. Thus, 22 studies of outcomes following FLP for TTTS were included in the analysis (Table 1).
These 22 studies included 2599 monochorionic twin pairs, of which outcome data were reported for 2447 twin pregnancies (Table 1). The published rate of TTTS recurrence following SFLP is 0 to 16%. Two of the largest studies (combined n = 451) included a minority of cases of non-selective FLP[16, 24]. Of the remaining 20 studies, there were 13 prospective observational studies, two prospective randomized trials and five retrospective studies (Table 1). The TTTS recurrence rate reported by the prospective studies ranged from 0 to 16% and that reported from retrospective studies ranged from 1 to 6%.
Among the studies that reported cases of TTTS recurrence (n = 17), the precise definition of ‘recurrent TTTS’ was available in 12 cases (70.6%). The definition was reported explicitly in nine studies and corresponding authors provided this information in another three cases. In most cases this was simply recurrence of the ‘twin oligohydramnios–polyhydramnios sequence’ (TOPS) at any stage following SFLP (Table 1). In two studies, a minimal interval of 1 week following laser therapy was mandated before a diagnosis of recurrent TTTS could be assigned[22, 31]. Another group included a requirement that the recurrence be diagnosed within 4 weeks of the primary SFLP procedure. A requirement for repeat intervention was included in the diagnostic criteria by one group.
Several reports included information on placental location[16-18, 24, 25, 27, 30-33]. However, sufficient data were not available to determine the impact of placental location on the risk of TTTS recurrence. One group compared outcomes following SFLP in anterior placentae with other placental locations, and found similar outcomes in both groups, including a similar rate of TTTS recurrence. Three other studies reported no difference in survival rates in cases of anterior versus posterior placenta[23, 24, 32].
Details on clinical management of recurrent TTTS were available for 71/108 (65.7%) cases of recurrent TTTS (Table 2). The most common management strategy was repeat SFLP (29.6%) followed by expectant management or delivery (19.4%). Perinatal outcome data specific to these cases of recurrent TTTS were poorly reported across the included studies (Table 3). Only three studies provided comprehensive outcome data for cases of recurrent TTTS[19, 27, 31]. Among 14 cases (n = 28 neonates) reported by Robyr et al., the rate of intact survival at 1 month of age was 50% (6/12) in the repeat SFLP group versus 37.5% (6/16) in the secondary amnioreduction group. In a small study by Ruano et al., there were three cases of TTTS recurrence following 19 SFLP procedures. All recurrences were treated with multiple amnioreductions per case and only 1/6 infants survived intact at 30 days (Table 3). In a third study (n = 18 neonates), the rate of intact neurological survival at 2 years was 55.6% (10/18), although information on clinical management of TTTS recurrence is not provided. Combining these three studies (n = 52 infants), the overall rate of intact survival was 44% (23/52; Table 3). Among liveborn twins, the incidence of death or neurologically impaired survival was 23.3% (7/30; Table 3). The data were inadequate to determine the impact of both TTTS recurrence and secondary therapeutic approach on perinatal outcomes, including gestational age at delivery.
Table 2. Clinical management of recurrent twin–twin transfusion syndrome: review of the literature
Finally, two additional studies examined the overall effect of SFLP complications on perinatal survival[18, 24]. On univariate analysis of 19/164 cases, Morris et al. found neither TTTS recurrence (P = 0.45) nor repeat SFLP (P = 0.99) to be associated with reduced perinatal survival at ≥ 28 days. In contrast, Habli et al. reported reduced survival rates in cases of ‘complicated’ SFLP compared with ‘uncomplicated’ SFLP. However, recurrent TTTS accounted for only a small proportion (3/59 (5.1%)) of ‘late complications’ (> 1 week following the procedure) in this study.
Since the publication of the Eurofetus randomized controlled trial in 2004, SFLP of placental vessels has become the treatment of choice for TTTS presenting at ≤ 26 weeks' gestation in MCDA twins[34, 35]. Although laser therapy has been shown to be superior to both amnioreduction[20, 30, 36] and conservative management for TTTS, it is not a risk-free intervention. Iatrogenic preterm premature rupture of membranes complicates up to 8% of diagnostic fetoscopies and up to 30% of operative fetoscopy procedures and is thus a significant concern after SLFP at ≤ 26 weeks' gestation[1, 38]. Women undergoing fetoscopic intervention for TTTS are also at risk of complications unique to the TTTS syndrome, including recurrent TTTS and TAPS[10, 11], which are probably different clinical manifestations of a common underlying pathophysiological mechanism. Finally, studies on the impact of operator experience on SFLP outcomes have yielded conflicting data[24, 39]. To date, the risk of recurrent TTTS after SFLP has been poorly, and inconsistently, quantified.
The present review finds a rate of recurrence of TTTS of 0 to 16% after a cumulative 2447 fetoscopic laser photocoagulation procedures. Recurrent TTTS is thought to occur as a consequence of incomplete coagulation of placental anastomoses during the laser procedure. Thus, it may be more accurately described as a persistence of the TTTS disease process, rather than a true recurrence. Published placental studies have examined the relationship between residual placental anastomoses and recurrent disease[40, 41]. Lopriore et al. reported residual anastomoses in one-third of 77 SFLP placentae on pathological examination. The majority of residual anastomoses were arteriovenous (AV), small (< 1 mm) and located near the placental margin. Interestingly, no case of recurrent TTTS was seen among the cases with residual placental anastomoses. However, a similar study by Lewi et al. did report an association between recurrence of TTTS and residual large AV anastomoses. In contrast, TAPS was more closely associated with missed smaller AV anastomoses.
No general consensus on the precise definition of recurrent TTTS exists. Most of the studies reviewed here used the diagnostic criteria from the original Eurofetus trial, that is polyhydramnios in the recipient (deepest vertical pool > 8 cm at ≤ 20 weeks' gestation or > 10 cm at > 20 weeks) and oligohydramnios in the donor (deepest vertical pool ≤ 2 cm). However, studies differ in the timing of recurrent TOPS. In a small proportion of included studies, the diagnosis of recurrent TTTS required a minimum of a 1-week interval after the SFLP procedure, implying that recurrent TTTS is, by definition a late complication[22, 31]. Several other studies did not mandate this interval in the definition of recurrence. A recent paper by Assaf et al. documented serial changes in amniotic fluid volume in the weeks following SFLP for TTTS. They found abnormal fluid levels in 30% of recipients and 20% of donors in the first postoperative week. The criteria for recurrent TTTS were met in 7% of cases in week 1, falling to 2% by week 2 without intervention. Their finding that fluid discordance in the early postoperative phase represented successful therapy in the majority of cases – with fluid volumes simply taking time to normalize – lends credence to the inclusion of a 1-week interval in the diagnosis of recurrent TTTS.
Fetoscopic laser photocoagulation is technically more challenging in the presence of an anterior placenta. An angled fetoscope is usually required to achieve appropriate access, and transplacental passage of the trocar may be unavoidable. Despite this, several studies reviewed here did not find anterior placental location to be associated with significantly different outcomes after SFLP from those with posterior placentae[23, 24, 32, 33]. However, very few studies have described the relationship between placental location and recurrent TTTS and, to our knowledge, no study has addressed the issue of placental location in secondary SFLP.
Recurrent TTTS has been described as ‘a very complex problem whose therapeutic approach is not yet defined’. Although this review provides clinically useful information on the risk of recurrence of TTTS following SFLP, several important clinical questions remain unanswered. Firstly, many studies of SFLP for TTTS do not report rates of disease recurrence, which would add to this growing body of literature[13, 42]. Secondly, in those studies that do incorporate recurrence data, the clinical management of such cases is not always clearly outlined. Additionally, outcome data such as gestational age at delivery, perinatal morbidity and neurodevelopmental parameters, are rarely stratified to allow identification of the recurrent cases.
In this review we found that recurrent TTTS appears to be associated with poor rates of neurologically intact survival (44% of all cases). This rate is lower than that for published series following a primary procedure[42-44]. A study of neurodevelopmental outcome at 2 years following primary SFLP reported normal development in 87% of the TTTS survivors following the procedure, comparable with outcomes in dichorionic twin controls delivered between 24 and 34 weeks' gestation. Indeed, a recent systematic review of perinatal outcomes after primary SFLP for TTTS reported an 8–18% rate of neurodevelopmental impairment among TTTS survivors treated with SFLP. Comparing liveborn twins only, this rate is lower than the 29% rate of neonatal death/impairment after recurrent TTTS reported in the present review. In contrast, Lopriore et al. found that disease recurrence did not predict neurological impairment among TTTS survivors, although numbers were small. However, data in the present review were insufficient to allow analysis of gestational age at delivery. It seems intuitive that twins complicated by ‘treatment-resistant’ TTTS or those involving multiple invasive procedures are at higher risk of preterm delivery. Thus, based on the available data, it is difficult to separate the disease (recurrent TTTS) from the effects of recurrent intervention and gestational age. Clearly the effect of TTTS recurrence and/or secondary fetoscopic intervention on perinatal outcome warrants further study.
Operative fetoscopic approaches aimed at minimizing the recurrence of TTTS are being investigated. The Dutch Solomon study to determine whether coagulation of the entire vascular equator reduces TTTS recurrence is currently recruiting (www.trialregister.nl; NTR1245). Nonetheless, several of the unanswered questions raised here have important clinical implications. Ideally, an expert consensus should be reached on the appropriate definition of ‘recurrent TTTS’ and whether a minimal interval between procedure and diagnosis is mandatory; this would facilitate better comparison across studies. Additionally, given that a proportion of SFLP procedures will be complicated by recurrent disease, improved data are needed to better inform clinicians on the optimal management of these complex cases. It is imperative that future studies report rates of TTTS recurrence and outline clearly both the management and outcome of cases of recurrent TTTS.
We are grateful to the authors of the original studies, Dr Enrico Lopriore, Dr Rodrigo Ruano, Dr Haruhiko Sago and Professor Alec Welsh, who provided us with additional data to supplement this review.